Update cookies preferences
 logo Idiap Research Institute        
 [BibTeX] [Marc21]
Leveraging Untranscribed Data for End-to-End Speech and Callsign Recognition in Air-Traffic Communication
Type of publication: Conference paper
Citation: Motlicek_SIDS2025_2025
Publication status: Published
Booktitle: SESAR Innovation Days 2025 (https://www.sesarju.eu/SIDS2025)
Year: 2025
Month: December
Location: Bled, Slovenia
Organization: Eurocontrol
URL: https://www.sesarju.eu/sites/d...
Abstract: Accurate Automatic Speech Recognition (ASR) and callsign recognition in Air Traffic Control (ATC) are vital for safety, yet conventional two-step systems rely on large amounts of manually transcribed data, which is both costly and limited. This paper introduces a practical alternative using TokenVerse, a unified end-to-end model trained under a dual-task framework and enhanced through semi-supervised learning. Our main contribution shows that the model can jointly learn callsign boundaries and speech recognition, improving performance on both tasks simultaneously. Additionally, by generating pseudo-labels for 500 hours of unlabeled audio, we substantially expand the effective training data. Experiments across multiple in-domain and out-of-domain ATC datasets demonstrate that the TokenVerse framework achieves state-of-the-art performance in both ASR and callsign detection, surpassing cascaded pipelines built on modern architectures (including Kaldi, XLSR/wav2vec 2.0, Zipformer, and Whisper). This work provides a robust and scalable foundation for deploying and continuously refining high-accuracy ATC systems in real-world settings where labeled data is inherently scarce. The end-to-end architecture is also relatively compact (approximately 317M parameters), making it well suited for real-time, low-latency deployment.
Main Research Program: Human-AI Teaming
Keywords: Air traffic control, Automatic Speech Recognition, semi-supervised learning
Projects: Idiap
armasuisse
Authors: Motlicek, Petr
Kumar, Shashi
Khalil, Driss
Prasad, Amrutha
Christof, Schüpbach
Added by: [UNK]
Total mark: 0
Attachments
    Notes